Definitions
For the purposes of this specification the term “pentenary alloy” refers to an alloy having 5 different elements. So for example, Cu(In,Ga)(S,Se)2 is a group IB-IIIA-VIA pentenary alloy wherein the 5 different elements are copper (Cu), indium (In), gallium (Ga), selenium (Se) and sulfur (S). Similarly the term “quaternary alloy” refers to an alloy having 4 different elements. So for example, Cu(In,Ga)Se2 is a group IB-IIIA-VIA quaternary alloy. Likewise, a ternary alloy has three different elements and a binary alloy has two different elements.
The term “homogeneous” alloy means that the different elements constituting the alloy are distributed homogeneously through the alloy such that the alloy has a substantially constant lattice parameter, interplanar spacing (referred to as d-spacing hereinafter) and band gap value throughout. In other words, the absolute shift of the main diffraction peak of the alloy [2θ(112)], characterised by glancing incident x-ray diffraction for glancing angles between 0.5° to 10°, is negligible.
Furthermore, for the purposes of this specification, a “heterogeneous” alloy means that the alloy includes a graded band gap structure and suffers from compositional grading such that one or more of the constituent elements of the alloy vary in atomic concentration through the alloy. The heterogeneous alloy may further include lattice mismatches in respect of the crystal structure and accordingly may suffer from a variation in the lattice parameters of the crystal structure through the alloy.
For the purposes of convenience, elements are referred to by their commonly accepted chemical symbols, including copper (Cu), indium (In), gallium (Ga), selenium (Se), sulphur (S), argon (Ar), molybdenum (Mo) and aluminium (Al). Also, the use of a hyphen (e.g. in Cu—In—Ga or Cu—In) does not necessarily indicate a compound, but indicates a coexisting mixture of the elements joined by the hyphen.
For the purposes of clarity, reference to group IB refers to the group in the periodic table comprising the elements of Cu, Ag and Au. Reference to group IIIA refers to the group in the periodic table comprising the elements B, Al, Ga, In and Ti. Furthermore, reference to group VIA refers to the group in the periodic table comprising the elements O, S, Se, Te and Po.
The use of a comma between two elements, for example (Se,S), (In,Ga) is merely used for the sake of convenience and so for example, (Se,S), is short hand for (Se1-ySy).
Semiconductor Film Material
Crystalline and multi-crystalline silicon is to date the primary material used in the production of solar modules/photovoltaic cells. The main problem associated with this material is the high cost of manufacturing. In an effort to reduce fabrication costs and increase material utilization, semiconductor thin film alloys have been the subject of intensive research. In this regard, group IB-IIIA-VIA alloys, such as CuInSe2, CuGaSe2 and CuInS2, are promising candidates for absorber layers in thin film photovoltaic cells or devices.
Of particular interest are semiconductor films comprising group IB-IIIA-VIA alloys wherein the alloy includes Ga in combination with another group III element, since the presence of Ga in such films results in semiconductor films with higher band gap values and subsequently, in solar/photovoltaic cell devices, with higher open-circuit voltages and reduced short circuit currents. Of even greater interest are semiconductor films comprising pentenary alloys (pentenary alloy semiconductor films).
In respect of semiconductor films comprising pentenary alloys having Cu(In1-xGax)(Se1-ySy)2 as a general formula, the band gap can be shifted systematically between 1.0 and 2.4 eV in order to achieve an optimum match with the solar spectrum. Optimization of this material system has already resulted in laboratory-scale solar cell devices with conversion efficiencies exceeding 18%.
Prior Art Processes
There are a number of methods for producing group IB-IIIA-VIA semiconductor films, the two most common methods being the traditional two step process and the co-evaporation process.
The Traditional Two Step Process
The above process typically involves (i) the deposition of metallic precursors such as Cu, In and Ga, on a substrate which is more often than not coated with molybdenum, by DC magnetron sputtering and then (ii) the reactive annealing of the precursors in an atmosphere containing Se and/or S vapours or H2Se/Ar and/or H2Se/Ar gases. These techniques are disclosed in an article by V. Alberts, J. H. Schön, and E. Bucher, Journal of Appl. Phys. 84(12), 1998, 6881 and by A. Gupta and S. Isomura, Sol. Energy Mater. Sol. Cells 53, 1998, 385.
Without wishing to be bound by theory and referring to an article by J. Palm, V. Probst, W Stetter and others, Thin Solid Films 451-452 (2004) 544-551, the selenisation of Cu—In—Ga metallic precursors is thought to produce binary alloys such as CuSe and In4Se3, Cu2-xSe and InSe. The subsequent reaction between these binary precursor phases at temperatures above 370° C. leads to the formation of the ternary alloy of CuInSe2 (CIS). It is believed that during selenisation, only the latter alloy is formed and the selenisation of Ga is kinetically impeded such that Ga is driven towards the molybdenum substrate during the formation of CIS. It is further believed that on further annealing, a separate layer of Cu(In,Ga)Se2 (CIGS) is formed such that a double layer structure results comprising a well crystallised CIS layer on top of a Ga-rich fine grained CIGS layer in contact with the back electrode. Extended annealing, which is commercially not preferable, results in Ga diffusion from the back electrode to the surface of the structure.
The effect of a segregated or graded film structure with most of the gallium residing at the back of the film, is that the absorber film exhibits a low band gap value in the active region of the photovoltaic cell, which ultimately limits the Voc of the device. (The open-circuit voltages (Voc) and short circuit currents (Jsc) of solar modules/photovoltaic cells are directly related to the band gap of the semiconductor material. In the case of CuInSe2 with a low band gap value of 1 eV, the Voc values are typically limited to 500 mV, while values close to 1000 mV can be achieved using a CuGaSe2 absorber layer with a higher band gap value of 1.65 eV.)
In addition, in the case of extreme grading, lattice mismatches within the graded absorber films introduce electrically active structural defects, which negatively impact on the device performance.
In order to overcome the disadvantage of a low band gap heterogeneous Cu(In,Ga)Se2 alloy, formed by the traditional two step process, films are commonly reacted with H2S.
Present industrial processes include a post-sulfurization step in which a certain fraction of the selenium species in the top surface region of the film are replaced with sulfur. (K. Kushiya, M. Tachiyuki, T. Kase, I. Sugiyama, Y. Nagoya, D. Okumura, M. Satoh, O. Yamase, and H. Takeshita, Sol. Energy Mater. Sol. Cells 49, 1997, 277; R. Gay, M. Dietrich, C. Fredric, C. Jensen, K. Knapp, D. Tarrant and D. Willett, Proceedings of the International Conference on E. C. Photovoltaic Solar Energy, Vol. 12(1), 1994, 935; and T. Nakada, H. Ohbo, T. Watanabe, H. Nakazawa, M. Matsui and A. Kunioka, Solar Energy Materials and Solar Cells 49, 1997, 285).
This approach ultimately results in the formation of a thin Cu(In,Ga)(Se,S)2 surface layer on the resultant graded Cu(In1-xGax)Se2 structure. The surface layer has an abrupt grading and the depth into the Cu(In,Ga)(Se2 structure is in the order of 50 nm.
The disadvantages of the above post-sulfurisation step, which is already applied on an industrial scale, are:    (i) the slow rate of exchange between the selenium and sulfur species in these films,    (ii) only a slight increase In the open-circuit voltages of solar cell devices are achieved,    (iii) high temperatures and long reaction times of between 90 to 120 minutes are required to achieve significant degrees of S incorporation, which ultimately increases the costs of the production process; and    (iv) the resulting alloys are heterogeneous, which prohibit effective control over the lattice parameters and band gap values.
It has also been demonstrated, in an article by M. Marudachalam, H. Hichri, R. Klenk, R. W. Birkmire, W. N. Schfarman and J. M. Schultz, Appl. Phys. Lett. 67(26), 1995, 3978, that Cu(In,Ga)Se2 thin films with improved homogeneity can be produced by the in-situ annealing of a phase-separated mixture of CuInSe2 and CuGaSe2 in argon in the temperature range of 500° C. to 600° C. for 60 to 120 minutes. However, Auger depth profiling of these specific alloys still revealed substantial variations in the In and Ga concentrations with depth, indicative of heterogeneous alloys.
In addition, the carrying out of the post-annealing step in an inert atmosphere resulted in substantial losses of Se from the film, which necessitated a second annealing step in H2Se/Ar. The additional post-annealing steps in an inert atmosphere as well as H2Se/Ar not only compromise the reproducibility of the process, but also make it commercially unviable.
Single Stage Techniques
In another attempt to produce homogeneous pentenary alloys, a complex single-stage technique has been developed. In this technique, disclosed in an article by I. M. Kötschau, H. Kerber, H. Wiesner, G. Hanna and H. W. Schock, Proceedings of the 16th European Photovoltaic Solar Energy Conference, 1-5 May 2000, Glasgow, UK, pp 724-727, all the elements (Cu, In, Ga, Se and S) are co-evaporated at constant fluxes in high vacuum from individual sources.
This technique allows for the controlled incorporation of gallium and sulfur into the films and hence in a decrease in the lattice parameters of the alloys. The subsequent increase in the band gap values of the pentenary alloys ultimately resulted in an increase in the open-circuit voltages of completed solar cell devices. However, glancing incident angle x-ray diffraction (GIXRD) at incident angles between 0.40 and 50 revealed a significant shift in the lattice parameters between the surface and the bulk of the material. The authors attributed this phenomenon to a copper depletion at the surface of the layer, which confirmed that the alloys were compositionally graded rather than homogeneous.
It has now surprisingly been found by the inventor that the significant problems discussed above can at least partially be overcome or reduced by controlling the formation of the ternary alloys in the selenization step such that the selenization reaction does not proceed to completion to form fully reacted ternary alloys in the absence of binary alloys.